This project concerns human B1 cells, and the immunoglobulin (Ig) produced by this B cell subpopulation. In recent years the success of B cell depletion therapies has raised the profile of B cells as key contributors to autoimmunity. However, the means by which B cells promote autoimmunity, and especially the precise origin of pathogenic autoantibodies, have not been fully worked out. For many years it has been suggested that B1 cells might be involved in lupus and related illnesses, but this has been impossible to evaluate because identifying markers for human B1 cells have been lacking. This gap has now been filled with our recent finding that human B1 cells co-express CD20, CD27 and CD43, and lack expression of CD70. With the phenotype of B1 cells in hand it is possible to examine the nature of B1 cell-generated Ig. Human B1 cells produce Ig with many features that suggest a relationship to autoimmunity. Human B1 cells produce anti-DNA autoantibody; human B1 cells express antibody that is somatically mutated; and, human B1 cell- derived antibody can be isotype-switched. And in lupus, not only is the proportion of B1 cells among all B cells increased 2-fold, but the fraction of B1 cells that express IgG is markedly increased 16-fold (to 35% of all B1 cells). Thus, human B1 cells that produce autoreactive Ig, somatically mutated Ig, and to a greater (lupus) or lesser (normal) extent, isotype-switched Ig, express the necessary attributes to be a source of autoreactive/mutated/switched autoantibody that is recognized as pathogenic. We hypothesize then that in lupus B1 cells, particularly IgG+ B1 cells, do, in fact, represent a source of pathogenic autoantibody. To address this hypothesis, we will study Ig produced by B1, naive, and memory B cells, comparing B cells from lupus patients with normal controls, and comparing IgG+ B1 cells with non-switched B1 cells. Specifically, 1) We will evaluate the structure of B1 cell Ig by single cell sorting, PCR amplification, and sequencing, focusing on VH/VD/VJ/VL usage, somatic mutation, N-region addition, and CDR3 length and charge; 2) We will evaluate the repertoire of B1 cell Ig by expression cloning, antibody purification, HEp-2 immunofluorescence, ELISA assay, and autoantigen array, focusing on the frequency of polyreactive, autoreactive antibodies; and, 3) We will evaluate the pathogenicity of B1 cell Ig by glomerular binding/immunofluorescence and histological examination of kidneys after murine in vivo autoantibody administration, focusing on the capacity of anti-DNA autoreactive/polyreactive antibodies to produce renal damage. We hypothesize that in lupus, B1 cells generate pathogenic, nephritogenic antibodies that are somatically mutated and isotype-switched, and thus represent the origin of at least some of the classical type of autoreactive Ig that has been thought to be purview of post-germinal center conventional B cells. The results of this work will lay the foundation for a new and discrete therapeutic target in lupus that may eliminate the need for wholesale destruction of all mature B cells.